[TRTLLM-14730][feat] Add image edit serving endpoint for visual generation models - #16987

Merged

yibinl-nvidia merged 14 commits into

NVIDIA:mainfrom

yibinl-nvidia:qwen-image-edit-openai-endpoint

Aug 18, 2026

Commits on Aug 17, 2026

Add OpenAI image edit endpoint for visual generation
yibinl-nvidia
committed
95d7ecd
Use server-side image edit capability check
yibinl-nvidia
committed
a49eb51
Add Qwen layered multiple image outputs
yibinl-nvidia
committed
caa6dac
Cover Qwen layered image edit response modes
yibinl-nvidia
committed
966165e
Address image edit review comments
yibinl-nvidia
committed
ce764e2
Address image edit endpoint review feedback
yibinl-nvidia
committed
394e973
Support Flux2 image edit serving for local paths
yibinl-nvidia
committed
d0d27de
Validate image edit input image bytes
yibinl-nvidia
committed
c782432
Address image edit review feedback
yibinl-nvidia
committed
a4178eb
Preserve auto edit dimensions for reference-sized models
yibinl-nvidia
committed
6cfa70e
Remove stale image edit capability tests
yibinl-nvidia
committed
76ea21f
Fix video seconds validation test setup
yibinl-nvidia
committed
763d830
skip threadcheck for current test
yibinl-nvidia
committed
c53c3eb
Remove CPU-only endpoint tests from B200 list
yibinl-nvidia
committed
4ce1940